32nd World Correspondence Champion is an American

Discussion of anything and everything relating to chess playing software and machines.

Moderator: Ras

CornfedForever
Posts: 650
Joined: Mon Jun 20, 2022 4:08 am
Full name: Brian D. Smith

Re: 32nd World Correspondence Champion is an American

Post by CornfedForever »

Ozymandias wrote: Mon Nov 14, 2022 10:59 am whatever solution is found...
Folks, there is no 'solution'.

It's like a swimming competition...only the lake has dried up.

Now competitors get to resort to whatever mode of transportation to get to the finish line. Just bury the past and call whatever you are doing...well, something else.
jefk
Posts: 1083
Joined: Sun Jul 25, 2010 10:07 pm
Location: the Netherlands
Full name: Jef Kaan

Re: 32nd World Correspondence Champion is an American

Post by jefk »

Bureaucracy is way more slow than engine progress, so whatever solution is found, it will surely arrive too late.
yep, there have been discussions here, more than one and
a half year ago, but i doubt if much of that has ever got through
to the ICCF (i.e. those who can make relevant decisions)
https://talkchess.com/forum3/viewtopic. ... e&start=60

Imho an ICCF task force could (or should) be setup
to investigate the problem and come-up with some
constructive and preferably simple improvements.
With some engine modifications such modifications could
even be tested in a Iccf tourn like simulation.

And if it's only about an alternative version (of the rules
or timing or whatever) compared to the current version, then
the question is 'why not' ? (except that it would take some time)
After a while, i guess many (eg 2200+ ranked) players would
like to give it a try, getting bored after all the draws in
the usual version. And after some experience with such an
alternative (imo improved) version, some final adaptations
could be made, to make it a real good (or even better)
variant, at least for correspondence chess.
My 2 cnts again as usual

would give it a try,
User avatar
towforce
Posts: 12709
Joined: Thu Mar 09, 2006 12:57 am
Location: Birmingham UK
Full name: Graham Laight

Re: 32nd World Correspondence Champion is an American

Post by towforce »

CornfedForever wrote: Mon Nov 14, 2022 2:51 pmFolks, there is no 'solution'.

It's like a swimming competition...only the lake has dried up.

Now competitors get to resort to whatever mode of transportation to get to the finish line. Just bury the past and call whatever you are doing...well, something else.

+1
Human chess is partly about tactics and strategy, but mostly about memory
User avatar
Ozymandias
Posts: 1537
Joined: Sun Oct 25, 2009 2:30 am

Re: 32nd World Correspondence Champion is an American

Post by Ozymandias »

jefk wrote: Mon Nov 14, 2022 2:55 pmthen the question is 'why not' ?
No reason needed, naysayers just will do their thing.
User avatar
towforce
Posts: 12709
Joined: Thu Mar 09, 2006 12:57 am
Location: Birmingham UK
Full name: Graham Laight

Re: 32nd World Correspondence Champion is an American

Post by towforce »

Ozymandias wrote: Mon Nov 14, 2022 10:19 pm
jefk wrote: Mon Nov 14, 2022 2:55 pmthen the question is 'why not' ?
No reason needed, naysayers just will do their thing.

That's a bit of a condemnation! :)

I have no objection to new forms of chess for correspondence chess - it just needs a couple of caveats:

* it won't be exactly the same game as it was before, and so getting agreement to the new rules will be tricky

* you're ultimately running a losing race against the march of computer chess: whatever new rules you come up with, chess software will probably adapt to them

* correspondence chess is never again going to be the same game that attracted so many people over so many centuries
Human chess is partly about tactics and strategy, but mostly about memory
smatovic
Posts: 3480
Joined: Wed Mar 10, 2010 10:18 pm
Location: Hamburg, Germany
Full name: Srdja Matovic

Re: 32nd World Correspondence Champion is an American

Post by smatovic »

towforce wrote: Mon Nov 14, 2022 6:22 pm
CornfedForever wrote: Mon Nov 14, 2022 2:51 pmFolks, there is no 'solution'.

It's like a swimming competition...only the lake has dried up.

Now competitors get to resort to whatever mode of transportation to get to the finish line. Just bury the past and call whatever you are doing...well, something else.

+1
+1
--
Srdja
jefk
Posts: 1083
Joined: Sun Jul 25, 2010 10:07 pm
Location: the Netherlands
Full name: Jef Kaan

Re: 32nd World Correspondence Champion is an American

Post by jefk »

towforce wrote:
it won't be exactly the same game as it was before, and so getting agreement to the new rules will be tricky
you're ultimately running a losing race against the march of computer chess: whatever new rules you come up with, chess software will probably adapt to them
1) agree, in other words no it won't be exactly the same,
but... so what ? FRC960 also is not the same btw..
For example, in TCEC they not use standard openings anymore, such
changes are inevitable (not sure what they use, they should
try to find unbalanced opening moves for computer chess
(there is an unbalanced opening book for humans i know,
not sure about computers).
2) you're ultimately running a losing race against the march of computer ?
Maybe but the main point for the time being IMO is mainly
to reduce the draw margin, which currently can be quite high
depending on endgame rules. Whereby still trying to
resemble rather closely conventional chess (eg opening setup).

PS as for cynical remarks about a dry swimming pool, i'm
not going to react, i'm not even a competitor in ICCF anymore
(almost for two years now) but it's general and constructive
interest in (computer) chess -rather than negativism- which
brought me to bring back this subject again (after eg Kaufman
and Arno Nickel almost two years ago. Selecting only a few)
openings out of all FRC960 (als Kaufman idea i think
(may also be an idea for correspondence chess, also not
saying as competitor, I've never played competitive FRC960
in fact. In conclusion, some evolution in some rules and
additional setups of chess are inevitable, also in
correspondence chess (although for ICCF changes will
probably go slowly, I agree); whether you like it or not.
Main point: resistance is futile
:)
jefk
Posts: 1083
Joined: Sun Jul 25, 2010 10:07 pm
Location: the Netherlands
Full name: Jef Kaan

Re: 32nd World Correspondence Champion is an American

Post by jefk »

most effective and simple rule change would be
not to allow a draw for 3x repetitions of a position anymore
In other words, zero points for the one first causing this
ie. not being able to prevent it.
(not sure if also something should be done about perpetual
check, here we have the 50 move rule).
So no i'll have a look at the latest WC32 final
and see if i can find an alternative winner
(with such alternative (imo better) rules
:)
PS of course this won't be perfect yet (later it
should become 1/4 instead of zero, and we have to
investigate if there's still a proper B/W balance
Anyway, you have to start somewhere don't you ?
2-2
Now it's ICCF's move
:)
PS already had a brief look, in game below the end
seems to be determined by 3 pos repetition.
White should have played better, as he cannot
prevent move/position repetition by Black at the end.
So hereby i declare mr Osipov the winner under
such alternative rules.
[Event "WC32/final"]
[Site "ICCF"]
[Date "2020.06.20"]
[Round "-"]
[White "Moll, Reinhard"]
[Black "Osipov, Sergey Adolfovich"]
[Result "1/2-1/2"]
[WhiteElo "2554"]
[BlackElo "2499"]
1.Nf3 {[%ccsnt 2020.07.03,13:35]} Nf6 {[%ccsnt 2020.07.03,14:14]}
2.c4 {[%ccsnt 2020.08.05,18:35]} e6 {[%ccsnt 2020.08.05,19:35]}
3.g3 {[%ccsnt 2020.08.13,16:13]} d5 {[%ccsnt 2020.08.13,17:04]}
4.d4 {[%ccsnt 2020.08.22,13:30]} Bb4+ {[%ccsnt 2020.08.22,17:00]}
5.Bd2 {[%ccsnt 2020.08.26,14:42]} Be7 {[%ccsnt 2020.08.26,15:18]}
6.Bg2 {[%ccsnt 2020.08.26,15:18]} O-O {[%ccsnt 2020.08.26,15:43]}
7.O-O {[%ccsnt 2020.08.26,18:48]} c6 {[%ccsnt 2020.08.26,20:56]}
8.Qc2 {[%ccsnt 2020.08.31,21:18]} b6 {[%ccsnt 2020.09.01,08:59]}
9.Rd1 {[%ccsnt 2020.09.03,16:52]} Nbd7 {[%ccsnt 2020.09.03,21:45]}
10.b3 {[%ccsnt 2020.09.04,10:30]} h6 {[%ccsnt 2020.09.04,12:32]}
11.Bc3 {[%ccsnt 2020.09.14,18:16]} Bb7 {[%ccsnt 2020.09.14,19:00]}
12.Nbd2 {[%ccsnt 2020.09.16,18:56]} Qc7 {[%ccsnt 2020.09.16,20:20]}
13.Rac1 {[%ccsnt 2020.09.17,17:51]} Rac8 {[%ccsnt 2020.09.17,21:40]}
14.Qb2 {[%ccsnt 2020.09.20,17:53]} Qb8 {[%ccsnt 2020.09.20,20:24]}
15.c5 {[%ccsnt 2020.09.23,11:52]} Ba6 {[%ccsnt 2020.09.23,18:32]}
16.Bf1 {[%ccsnt 2020.09.24,13:49]} Ne4 {[%ccsnt 2020.09.24,16:05]}
17.Nxe4 {[%ccsnt 2020.09.27,08:25]} dxe4 {[%ccsnt 2020.09.27,08:40]}
18.Ne1 {[%ccsnt 2020.09.27,08:40]} f5 {[%ccsnt 2020.09.27,09:40]}
19.cxb6 {[%ccsnt 2020.10.05,15:05]} Nf6 {[%ccsnt 2020.10.05,17:41]}
20.bxa7 {[%ccsnt 2020.10.06,07:05]} Qxa7 {[%ccsnt 2020.10.06,07:43]}
21.Ng2 {[%ccsnt 2020.10.06,07:43]} Nd5 {[%ccsnt 2020.10.06,08:27]}
22.Bd2 {[%ccsnt 2020.10.18,22:06]} Bf6 {[%ccsnt 2020.10.19,06:58]}
23.Qa3 {[%ccsnt 2020.10.30,12:47]} Bxd4 {[%ccsnt 2020.10.30,15:19]}
24.Be1 {[%ccsnt 2020.11.01,09:15]} e5 {[%ccsnt 2020.11.01,14:08]}
25.e3 {[%ccsnt 2020.11.01,14:08]} Bxf1 {[%ccsnt 2020.11.01,14:12]}
26.Qxa7 {[%ccsnt 2020.11.02,05:54]} Bxa7 {[%ccsnt 2020.11.02,06:32]}
27.Kxf1 {[%ccsnt 2020.11.02,06:32]} g5 {[%ccsnt 2020.11.02,07:30]}
28.Rc4 {[%ccsnt 2020.11.04,08:57]} Nb6 {[%ccsnt 2020.11.04,14:11]}
29.Rc2 {[%ccsnt 2020.11.09,11:31]} Nd5 {[%ccsnt 2020.11.09,18:47]}
30.Rc4 {[%ccsnt 2020.11.14,09:34]} Nb6 {[%ccsnt 2020.11.14,12:39]}
User avatar
Ozymandias
Posts: 1537
Joined: Sun Oct 25, 2009 2:30 am

Re: 32nd World Correspondence Champion is an American

Post by Ozymandias »

jefk wrote: Tue Nov 15, 2022 2:30 pm in game below the end
seems to be determined by 3 pos repetition.
White should have played better, as he cannot
prevent move/position repetition by Black at the end.
So hereby i declare mr Osipov the winner under
such alternative rules.
If white replies to Nd5 with Rc4 (a third time), yes, he shouldn't score as well as black. Such a spineless game.
jefk
Posts: 1083
Joined: Sun Jul 25, 2010 10:07 pm
Location: the Netherlands
Full name: Jef Kaan

Re: 32nd World Correspondence Champion is an American

Post by jefk »

spineless game ? maybe,

NB some additional comments;
1) of course my comment of declaring Osipov
a (alternative) winner was a joke because
2) once such adapted rules are in place ofcourse
the player who will possibly encounter disadvantage
under such rules, can try to avoid it, and
3) meanwhile if found more with 3 mover
repetition, but haven't gone through
the whole tourn pgn (yet?)

[Event "WC32/final"]
[White "Ulbig, Stefan"]
[Black "Moll, Reinhard"]
[Result "1/2-1/2"]

[Event "WC32/final"]
[White "Zajontz, Rainer"]
[Black "Moll, Reinhard"]
[Result "1/2-1/2"]

Main point is, the old Kaufman remark that
one of the most effective methods of reducing
the drawing rate is doing something about the
3 pos repetition, indeed may well be true at least
looking at the not infrequent occurrence of such games.
How to exactly penalty such occurrence isn't
still completely clear for me (eg the 1/4 3/4 suggestion
by Kaufman). Once such a rule is in place ofcourse
it can be avoided, but... apparently (as was already stated
in the earlier talkchess discussion about corresp chess with Nnue)
someone had done a simulation (adapted SF engine !) and
indeed found that such new rules would reduce the
draw rate by a significant amount.

So i don't think i'm talking rubbish here and may be
on to something (useful); despite the naysayers ofcourse
:)
But then maybe help may come in this forum with the
best comp chess expert in the world (not necessarily ICCF)
and meanwhile i'll go and read the original kaufman proposal
and the bloke who had done a simulation (0.0 instead of 1/4
may be a bit drastic, ofcourse, but also with this a simulation
could be done, eg even in an engine tourn (eg 5 mins time
per move or so, as first estimate).